移动GPU中通用图形应用编程接口库的设计
|
摘要
随着移动设备的快速发展和图形硬件加速器种类的与日俱增,图形API库的灵活性和高效性显得尤为重要。传统图形应用程序开发是基于特定的图形API标准在其专用图形处理器上实现加速渲染,但这会对图像的复杂度和高清化造成一定的局限性。因此,基于自主设计的移动GPU硬件架构,通过分析OpenGL ES 2.0和OpenVG 1.1两种图形库标准,设计了一种包含通用管理层、函数容错预处理层和命令解析层的3层抽象结构的通用图形应用编程接口库。在Linaro操作系统和ZC706 FPGA搭建的软件平台上完成对所设计通用图形API库的验证。结果表明,所设计的通用图形API库符合以上两种图形库标准,可以完成基本图形的渲染操作。
With the rapid development of mobile devices and the increasing variety of graphics hardware accelerators, the flexibility and efficiency of the graphics API library are important. Traditional graphics application development is based on specific graphics API standards to achieve accelerated rendering on its dedicated graphics processor, but this will impose limitations on the complexity and high-definition of the image. Therefore, this paper designs an API library based on the independently designed mobile GPU hardware architecture for general graphics. By analyzing the two library standards of OpenGL ES 2.0 and OpenVG 1.1, an API library for general graphics is designed by using a three-layer abstract structure including general management layer, function preprocessing layer and function fault pre-parsing layer. The API library for general graphics is validated based on the Linaro OS and ZC706 FPGA. The experimental results show that the designed API library for general graphics conforms to the API library, and completes the basic graphics rendering operation.
With the rapid development of mobile devices and the increasing variety of graphics hardware accelerators, the flexibility and efficiency of the graphics API library are important. Traditional graphics application development is based on specific graphics API standards to achieve accelerated rendering on its dedicated graphics processor, but this will impose limitations on the complexity and high-definition of the image. Therefore, this paper designs an API library based on the independently designed mobile GPU hardware architecture for general graphics. By analyzing the two library standards of OpenGL ES 2.0 and OpenVG 1.1, an API library for general graphics is designed by using a three-layer abstract structure including general management layer, function preprocessing layer and function fault pre-parsing layer. The API library for general graphics is validated based on the Linaro OS and ZC706 FPGA. The experimental results show that the designed API library for general graphics conforms to the API library, and completes the basic graphics rendering operation.
引文
[1] Hsiao S F, Li S Y, Tsao K H. Low-power and high-performance design of OpenGL ES 2.0 graphics processing unit for mobile applications[C]// IEEE International Conference on Digital Signal Processing. Singapre: IEEE, 2015:110-114.
[2] Chen S H, Lin H M, Wei H W, et al. Hardware/software co-designed accelerator for vector graphics applications[C]// Application Specific Processors. San Diego CA: IEEE, 2011:108-114.
[3] 刘晖, 田泽, 黎小玉,等. 3 D图形处理器API符合性验证方法关键技术研究[J]. 计算机技术与发展, 2014(10):193-196.
[4] Lee H, Baek N, Lee I, et al. Accelerating OpenVG and SVGTiny with multimedia hardware[C]// IEEE International Conference on Consumer Electronics. LAS Vegas, V V: IEEE, 2011:917-918.
[5] Lee I, Lee H, Baek N. Research on Implementation of Graphics Standards Using Other Graphics API’s[J]. 2011, 262:197-202.
[6] Yoon J, Baek N, Lee H. ARINC661 Graphics Rendering Based on OpenVG[C]// International Conference on It Convergence and Security. Kuala Lumpur: IEEE, 2015:1-4.
[7] Chen S H, Lin H M, Hsieh C C, et al. TurboVG: A HW/SW co-designed multi-core OpenVG accelerator for vector graphics applications with embedded power profiler[C]// Asia and South Pacific Design Automation Conference. Yokohama: IEEE Press, 2011:97-98.
[8] 余久久, 傅廷亮, 张佑生. 基于OpenGL的雪景视景仿真设计与实现[J]. 实验室研究与探索, 2014, 33(9):105-110.
[9] Abdallah Y, Abdelhamid A, Elarif T, et al. Comparison between OpenGL ES and metal API in medical volume visualisation[C]// IEEE Seventh International Conference on Intelligent Computing and Information Systems, Cairo: IEEE, 2016:156-160.
[10] Oh A, Sung H, Lee H, et al. Implementation of OpenVG 1.0 using OpenGL ES[C]// Conference on Human-Computer Interaction with Mobile Devices and Services, Mobile Hci 2007, Singapore, ACM, 2007:326-328.
[11] Robart M. OpenVG paint subsystem over openGL ES shaders[C]// International Conference on Consumer Electronics. Las Vegas, NV: IEEE, 2009:1-2.
[12] 谢晓燕, 崔继兴, 邓军勇,等. 基于VxWorks图形应用编程接口库的设计与实现[J]. 西安邮电大学学报, 2016, 21(2):98-100.
[13] Crockett L H, Elliot R A, Enderwitz M A, et al. The Zynq Book: Embedded Processing with the Arm Cortex-A9 on the Xilinx Zynq-7000 All Programmable Soc[M]. Strathclyde Academic Media, 2014.
[14] 马城城, 田泽, 黎小玉. 基于GPU FPGA芯片原型的VxWorks下驱动软件开发[J]. 计算机技术与发展, 2013, 23(7):84-86.
[15] Kadav A, Swift M M. Understanding modern device drivers[J]. Acm Sigarch Computer Architecture News, 2012, 40(1):87-98.
[16] 孙文华, 黄凌云. 嵌入式Linux驱动开发技术综述[J]. 广东通信技术, 2015, 35(11):68-73.
[17] 谭显强. 基于FPGA的3D图形处理器IP核的设计与实现[D]. 南京:南京航空航天大学, 2010.
[2] Chen S H, Lin H M, Wei H W, et al. Hardware/software co-designed accelerator for vector graphics applications[C]// Application Specific Processors. San Diego CA: IEEE, 2011:108-114.
[3] 刘晖, 田泽, 黎小玉,等. 3 D图形处理器API符合性验证方法关键技术研究[J]. 计算机技术与发展, 2014(10):193-196.
[4] Lee H, Baek N, Lee I, et al. Accelerating OpenVG and SVGTiny with multimedia hardware[C]// IEEE International Conference on Consumer Electronics. LAS Vegas, V V: IEEE, 2011:917-918.
[5] Lee I, Lee H, Baek N. Research on Implementation of Graphics Standards Using Other Graphics API’s[J]. 2011, 262:197-202.
[6] Yoon J, Baek N, Lee H. ARINC661 Graphics Rendering Based on OpenVG[C]// International Conference on It Convergence and Security. Kuala Lumpur: IEEE, 2015:1-4.
[7] Chen S H, Lin H M, Hsieh C C, et al. TurboVG: A HW/SW co-designed multi-core OpenVG accelerator for vector graphics applications with embedded power profiler[C]// Asia and South Pacific Design Automation Conference. Yokohama: IEEE Press, 2011:97-98.
[8] 余久久, 傅廷亮, 张佑生. 基于OpenGL的雪景视景仿真设计与实现[J]. 实验室研究与探索, 2014, 33(9):105-110.
[9] Abdallah Y, Abdelhamid A, Elarif T, et al. Comparison between OpenGL ES and metal API in medical volume visualisation[C]// IEEE Seventh International Conference on Intelligent Computing and Information Systems, Cairo: IEEE, 2016:156-160.
[10] Oh A, Sung H, Lee H, et al. Implementation of OpenVG 1.0 using OpenGL ES[C]// Conference on Human-Computer Interaction with Mobile Devices and Services, Mobile Hci 2007, Singapore, ACM, 2007:326-328.
[11] Robart M. OpenVG paint subsystem over openGL ES shaders[C]// International Conference on Consumer Electronics. Las Vegas, NV: IEEE, 2009:1-2.
[12] 谢晓燕, 崔继兴, 邓军勇,等. 基于VxWorks图形应用编程接口库的设计与实现[J]. 西安邮电大学学报, 2016, 21(2):98-100.
[13] Crockett L H, Elliot R A, Enderwitz M A, et al. The Zynq Book: Embedded Processing with the Arm Cortex-A9 on the Xilinx Zynq-7000 All Programmable Soc[M]. Strathclyde Academic Media, 2014.
[14] 马城城, 田泽, 黎小玉. 基于GPU FPGA芯片原型的VxWorks下驱动软件开发[J]. 计算机技术与发展, 2013, 23(7):84-86.
[15] Kadav A, Swift M M. Understanding modern device drivers[J]. Acm Sigarch Computer Architecture News, 2012, 40(1):87-98.
[16] 孙文华, 黄凌云. 嵌入式Linux驱动开发技术综述[J]. 广东通信技术, 2015, 35(11):68-73.
[17] 谭显强. 基于FPGA的3D图形处理器IP核的设计与实现[D]. 南京:南京航空航天大学, 2010.